AI voice cloning technology has advanced to the point where it can mimic human voices with high accuracy, making it a growing concern for fraud. Scammers are using these synthetic voices to trick people into believing they are speaking with a loved one in distress. One X user recently received a call from someone claiming to be their wife, who supposedly needed help with a forgotten wallet and asked for her credit card. However, the user was aware their wife was at home and drove a Tesla. They noticed the voice was familiar but had an odd rhythm, leading them to suspect it was a deepfake or voice filter. Although they did not fall for the scam, the incident highlighted how convincing these imitations can be. The story gained attention on Reddit, where users shared mixed reactions. Some were doubtful, while others described similar experiences, such as being told that family members were in trouble and needed bail money. Research supports the growing concern. A 2026 study titled "Evaluating AI Models' Capability to Automate Voice Phishing Attacks" by Fred Heiding and colleagues found that up to 36.1% of 4,100 U.S. adults said they would comply with or were unsure whether they would comply with an AI-generated "relative in distress" scam. The researchers clarified that this figure reflects self-reported willingness, not the number of people who were actually scammed. The study also revealed that the persuasiveness of the message mattered more than how realistic the voice sounded. People who frequently used AI were not better at identifying synthetic voices than those with no AI experience. While scams involving fake emergencies from loved ones are not new, AI has made them more dangerous. If a scammer sounds exactly like a family member, it becomes extremely difficult to determine who is actually on the other end of the call. To combat this, experts recommend setting up a family code word that everyone knows, which a scammer would be unlikely to guess. Another precaution is to hang up and call the person back using a known number. If a family member suddenly calls asking for urgent money, do not rely on the voice alone. Instead, end the call and contact them directly. The effectiveness of personal verification was demonstrated in a 2024 incident involving a Ferrari executive. After receiving suspicious messages and a phone call impersonating CEO Benedetto Vigna, the executive asked the caller to identify a book the real CEO had recently recommended. The caller could not answer, exposing the scam. Mark Beare, Head of Consumer at Malwarebytes, emphasized that AI's impact is not just on technology but on people's trust. He noted that when people can no longer trust what they see, hear, or who they’re talking to, the consequences extend beyond individual scams. A Malwarebytes survey from June 2024 found that 85% of people said it’s hard to tell a scam from the real thing, up from 66% the previous year. Additionally, 88% said it’s becoming increasingly difficult to determine which online content is genuinely human or real.